The SaaS paid acquisition playbook.

Most paid media advice is ecommerce advice wearing a neutral costume. It assumes the conversion is the revenue event, the feedback loop is days long, and the value of a customer is known at purchase. SaaS violates all three: the conversion is a trial or a demo request worth nothing by itself, the feedback loop runs through a sales cycle measured in weeks or quarters, and the value of a customer unfolds over years of retention you cannot see on day one.

But the assumption that does the most damage is subtler than any of those, and it is this: that if the unit economics work, the spend is affordable. In ecommerce that is roughly true, because the cash comes back this week. In SaaS the money goes out today and returns over months, so the binding constraint on how much you can spend is not whether a customer is profitable. It is when they pay. That makes SaaS paid acquisition a cash-flow problem before it is a marketing problem, and the playbook has to start there.

Payback period is the growth speed limit

LTV:CAC is a solvency ratio: it tells you whether a customer is worth acquiring. Payback period is a liquidity constraint: it tells you how long your cash is gone. Companies rarely fail because their customers were unprofitable. They fail because they ran out of money while being profitable on paper, which is a liquidity failure that no solvency ratio can warn you about.

Here is the arithmetic that matters, for a product at $2,000 MRR, 80% gross margin, and a $6,000 CAC:

Monthly contribution per customer  $2,000 x 0.80 = $1,600
CAC                                $6,000
Payback                            6,000 / 1,600 = 3.75 months

Spend $60,000/month -> 10 new customers/month
Each cohort adds $16,000/month of contribution

Month 1  inflow $16,000   vs  spend $60,000
Month 2  inflow $32,000   vs  spend $60,000
Month 3  inflow $48,000   vs  spend $60,000
Month 4  inflow $64,000   vs  spend $60,000  <- self-funding

Cumulative cash required to reach that point: ~$120,000

The month the program starts funding itself is the payback period. That is not a coincidence — it is the same number viewed from the treasury instead of from the cohort. And it means every increment of spend has a price tag beyond the spend itself: raising the monthly budget by $60,000 requires roughly $120,000 of working capital to carry the increment until it turns over. An acquisition plan that ignores this is not a plan, it is a hope about the bank balance.

Billing terms move the budget more than campaigns do

The consequence of the above is uncomfortable for anyone whose job is optimizing accounts: the largest available lever on your affordable acquisition budget is usually not in the ad account at all. Take two companies with identical products, identical CAC, and identical LTV:CAC:

A: billed monthly
   $24,000 ACV -> $1,600/month contribution
   payback 3.75 months, ~$120k working capital per $60k step

B: billed annually up front, 15% prepay discount
   $20,400 collected on day 0 -> $16,320 contribution at once
   payback immediate; the customer funds the next acquisition

Cost of the discount: ~$2,900 of contribution per customer
Benefit: the working capital requirement goes to roughly zero

Company B can scale acquisition as fast as it can find customers. Company A can scale as fast as its balance sheet allows, which at any given moment is a much smaller number. The prepay discount looks like a pricing decision and is in fact the single biggest acquisition decision on the list — a 15% giveaway that buys an unbounded growth rate. Meanwhile the campaign optimization everyone is arguing about moves efficiency by a few percent.

That trade is not universally correct. A well-capitalized company should not give away 15% of revenue to solve a cash problem it does not have, and doing so permanently lowers realized LTV in exchange for a constraint it was never bound by. But a runway-constrained company that has never modeled this is leaving the largest lever it owns untouched while asking its media buyer for another 5% on CPA. Whether it applies to you is a function of the balance sheet, which is exactly why this belongs in an acquisition playbook rather than a finance one.

Measurement first: the pipeline feedback loop

With the cash constraint understood, the highest-leverage build is not a campaign — it is closing the loop between the CRM and the ad platforms. Offline conversion import, pushing qualified-opportunity and closed-won events with their values back into Google and Meta, changes what the bidding algorithms optimize toward: from people who fill forms to people who become pipeline. Every transformed SaaS account we have worked on had this plumbing; every account stuck at mediocre had algorithms diligently maximizing a form-fill count nobody in the revenue meeting cared about. Build it before scaling anything.

Keywords by intent tier, budgets to match

• →Category terms ("crm software", "expense management tool") — the demand you build the account around. Expensive, competitive, and worth it when your close rates support the CPC math.
• →Competitor and alternative terms ("[rival] alternative", "[rival] pricing") — the highest-intent non-brand traffic in SaaS. These searchers have a budget and a shortlist; comparison landing pages, not the homepage, do the converting.
• →Problem terms ("how to track billable hours") — cheaper, earlier, and the natural budget for capturing buyers before they know the category exists. Expect longer conversion lag and measure accordingly.
• →Brand — protect it if competitors bid on you, and report it separately, always. Blended SaaS numbers flatter even harder than ecommerce ones because brand converts at rates category terms never will.

Trial vs demo changes the whole account

Self-serve trials and sales-led demos are different acquisition economics, and the account structure should say so. Trial motions produce volume with weak per-event value — the game is qualifying the click before the signup (pricing transparency, honest feature framing) and optimizing to activation milestones, not signups. Demo motions produce scarce, expensive, high-variance events — the game is lead quality above all, with tight ICP framing in the copy doing sales-development work at the auction stage. Hybrid companies should run the motions as separate campaigns with separate targets; blending them hands the algorithm a value function neither motion believes in.

The two motions also have different cash profiles, which is the version of this decision the cash constraint above makes visible. Trials are cheap per event and slow to revenue; demos are expensive per event and faster to a signed contract, often an annually-billed one. A runway-constrained company frequently finds that the demo motion is affordable and the trial motion is not, despite the trial motion having the better-looking CPA — because CPA measures the wrong thing when the constraint is working capital.

The channel order that compounds

Google Search on high-intent tiers first — capture beats creation while you can afford it. LinkedIn second for genuinely ICP-gated B2B, where its targeting justifies CPCs that look absurd next to Google until you weigh lead quality. Meta and YouTube third, as the audience layer — retargeting the research-cycle audience and prospecting lookalikes seeded from closed-won customers, not trial signups. Microsoft Ads as the quiet fourth, importing your proven Google structure for a work-hours desktop audience that fits B2B better than most teams expect.

Each addition is funded by the proven economics of the layer before it, and each is judged on pipeline contribution over a full sales cycle rather than on platform-reported CPA in week two. Note what the ordering is actually optimizing: not reach, but the speed at which each layer returns cash to fund the next. Channels that convert existing demand pay back faster than channels that create it, which is why capture comes first even for companies whose long-term growth depends on demand creation.

Reviewing a SaaS month without fooling yourself

A pre-and-post comparison around a change is correlational rather than causal in any channel — other campaigns move, seasonality moves, platform algorithms move. In SaaS there is a second, worse problem layered on top: the pipeline you are looking at this month was largely created by decisions made one or two sales cycles ago. So a month-end review that credits this month's changes with this month's pipeline is not merely overclaiming causality; it is attributing results to the wrong quarter entirely.

Three disciplines keep the review honest. State what moved, over what window, and how confident you are — never that a change caused an outcome. Report leading indicators (lead quality, opportunity rate, cost per qualified opportunity) against the current period and lagging indicators (closed-won, CAC payback) against the period that actually produced them. And if the company has no written acquisition targets, say so as a finding rather than inventing a benchmark to compare against — an unstated target is not a neutral condition, it is the reason nobody can agree whether the month was good.

The failure mode to design against

SaaS paid acquisition fails slowly and politely: lead volume healthy, CPL acceptable, dashboards green, pipeline quietly starving. The defense is cultural as much as technical — marketing and sales looking at one funnel with one definition of qualified, cost-per-opportunity and cost-per-closed-won reported beside CPL, and unit economics that specify what a lead is actually allowed to cost. When the numbers that matter are on the same page as the numbers that flatter, the flattering ones lose their power to steer the account wrong.

Where this model breaks down

• →The payback arithmetic above ignores churn, which is defensible over a 3-4 month window and indefensible over a year. If monthly churn is above roughly 2%, the self-funding month arrives later than the simple calculation says, because part of each cohort stops paying before the program turns over.
• →It also assumes a stable CAC as spend rises, which is false — scaling buys colder demand, so the payback period lengthens as the budget grows. The honest version of the model recalculates payback at the new CAC after each step rather than holding the original figure.
• →Contribution margin here uses gross margin, which for most SaaS overstates it. Support, onboarding and success costs scale with customers and belong in the denominator if you want the number to survive a CFO reading it.
• →Below roughly 20 to 30 closed deals per quarter from paid, close rates are too noisy to set targets from. Cost per qualified opportunity is the most reliable metric available at that volume; cost per closed-won is a number you can compute but should not steer by.
• →None of this tells you whether the offer is right. A cash-efficient acquisition program pointed at a product the market does not want will simply reach the wrong conclusion faster and cheaper.

The playbook compresses to this: understand the cash constraint before setting a budget, fix billing terms if that is where the real lever is, close the CRM loop before scaling, tier keywords by intent and fund them by close-rate math, split trial and demo economics, add channels in payback order, and report pipeline beside leads with the right period attached to each. None of it is exotic — it is the ecommerce discipline translated into an industry where the revenue event hides months behind the click and the bank account notices before the dashboard does. It is also, word for word, how we run Google Ads for SaaS clients.